Gestational trophoblastic disease includes a spectrum of diseases with varying propensity for local invasion and metastasis including partial and complete hydatidiform moles (partial molse and complete mole), placental site trophoblastic tumor, and choriocarcinoma. 1–5 Both complete and partial moles may develop persistent gestational trophoblastic tumor with local uterine invasion and dissemination. 1–7 Importantly, gestational trophoblastic tumor is remarkably curable with chemotherapy, even with the occurrence of multiple metastases. Furthermore, after attaining complete remission with chemotherapy, patients may preserve their fertility and still have a normal future reproductive outcome. 8–11
All gestational trophoblastic disease produces human chorionic gonadotropin (hCG), which can be measured in both the serum and the urine. The serum hCG level is a sensitive indicator for following the disease process, including treatment response and detection of persistent tumor and relapse. After molar evacuation patients are generally followed up with serial hCG values to assure the attainment of complete sustained remission. The American College of Obstetricians and Gynecologists 12 suggests that patients with hydatidiform mole should obtain hCG levels 48 hours after evacuation and every 1 to 2 weeks until levels are undetectable. Once undetectable levels are attained, follow-up measurements are made at 1–2-month intervals for an additional 6–12 months. 12 Although current recommendations for hCG follow-up safely ensure remission, the advised monitoring is often not completed. Investigators at the University of Southern California reported that 19% of their patients with molar pregnancy did not complete their gonadotropin follow-up. 13 Additionally, studies from Korea and Chicago noted that 27% and 81% of patients with molar pregnancy, respectively, prematurely discontinued the recommended hCG monitoring. 14,15
In this study we analyzed hCG follow-up data from 400 patients treated at the New England Trophoblastic Disease Center for complete or partial hydatidiform mole. We wanted to determine how often patients did not complete the advised follow-up and to identify possible factors associated with the failure to complete hCG monitoring. This study also sought to determine how often patients with molar pregnancy who do not complete follow-up relapse after attaining at least one undetectable hCG value.
MATERIALS AND METHODS
Data were collected using the New England Trophoblastic Disease Center database, which contains the data of patients with gestational trophoblastic disease from June 1, 1965 to the present. Beginning in 1975, when the β-hCG radioimmunoassay was available at our institution, every fourth patient with molar pregnancy was selected as a participant in the study until a study population of 400 patients was accrued. Patient information regarding medical history, obstetric history, social history, details of their presenting gestational trophoblastic disease, treatment, the staging and the histological diagnoses, and serum hCG values was collected from this database. This study was approved by the Institutional Review Board at our institution.
Four hundred patients with molar pregnancy were randomly selected to analyze the above data and hCG levels during follow-up. All patients were observed at our center from the time of evacuation until either remission or persistence. If persistence was documented, treatment was administered at our center. The following demographic factors were determined in each patient: age, gravidity, parity, marital status, insurance type, and distance from patient residence to trophoblast center. All patients selected had to obtain at least three serum hCG levels after molar evacuation to be included in the study.
The following recommendations for molar pregnancy follow-up are made to our patients at the New England Trophoblastic Disease Center: All patients with molar pregnancy should obtain serum hCG levels weekly until undetectable for 3 consecutive weeks. Patients with partial mole who attain undetectable hCG levels within 7 weeks after evacuation are then followed up with monthly hCG levels until undetectable for 3 months. The remaining patients with partial mole as well as all patients with complete mole are followed up with monthly hCG values until undetectable for 6 months. Patients who miss one or two serum hCG levels are telephoned by a physician or nurse and their primary physician is also contacted to ascertain if any hCG data had been collected through his or her office. All patients and their primary physicians are also regularly contacted to determine whether any tumor relapse is detected. If there is continued lack of follow-up, a registered letter is sent to the patients warning them of potential serious risks to their health if they neglect to complete their follow-up. All patients are provided a patient information booklet written in simple English or Spanish, which explains the risks of molar pregnancy and the extreme importance of reliable contraception and careful follow-up.
The statistical significance of data was evaluated by using the Student two-tailed t and Pearson χ2 tests. The level of significance was assigned at P < .05. All analyses were performed using SPSS (SPSS Inc., Chicago, IL) 10.0 and Excel 97 (Microsoft Corp., Redmond, WA).
Patient ages ranged from 13 to 54 years, with a mean age of 26.5 (± standard deviation 6.2). The histological diagnosis was complete mole in 293 cases (73%) and partial mole in 107 cases (27%).
Among the 400 patients with molar pregnancy, 320 (80%) achieved remission, 67 (16.8%) developed persistent gestational trophoblastic tumor, and 13 (3.2%) were lost to follow-up before attaining an undetectable hCG value.
Table 1 summarizes hCG follow-up in complete and partial molar pregnancies. Complete molar pregnancy was diagnosed in 293 cases (73%), and 62 of these patients (21%) developed persistent gestational trophoblastic tumor. The mean preevacuation hCG value in patients with complete mole was 248,817 IU/L. The mean preevacuation hCG level in patients who achieved remission or developed persistent gestational trophoblastic tumor was 195,826 IU/L or 423,688 IU/L, respectively (P = .001). Among patients who achieved spontaneous remission, the mean time to attain an undetectable hCG value was 8.4 weeks.
Partial molar pregnancy was diagnosed in 107 cases (27%), and five of these patients (4.7%) developed persistent gestational trophoblastic tumor. The mean preevacuation hCG value in patients with partial mole was 80,657 IU/L. The mean preevacuation hCG level in patients who achieved remission or developed persistent gestational trophoblastic tumor was 75,411 IU/L or 259,000 IU/L, respectively. Because of the small number of patients with persistent gestational trophoblastic tumor after partial mole, statistical significance in differences between preevacuation hCG levels could not be evaluated. Among patients who achieved spontaneous remission after partial mole, the mean time to attain an undetectable hCG value was 5.8 weeks.
There were significant differences between complete and partial mole in preevacuation serum hCG levels (P = .02) and in the number of weeks until hCG became undetectable (P < .001). There was no significant difference in the gestational age at the time of presentation and subsequent evacuation (mean of 13.5 weeks of amenorrhea for complete mole and 13.9 weeks for partial mole). In 36.6% of complete mole and 24% of partial mole the mole was the first pregnancy for the patient. The frequency of prior mole was 3% in complete mole and 4% in partial mole.
Among the 333 patients (231 cases of complete mole and 102 cases of partial mole) who did not develop persistent gestational trophoblastic tumor, only 211 patients (63%) completed the recommended hCG follow-up (Figure 1). Thirteen patients (4%) were lost to follow-up without attaining even one undetectable hCG value. One hundred nine patients (33%) achieved undetectable hCG values but did not complete their entire follow-up. In eight patients (2.4%) the reason for discontinuing follow-up was their conceiving before the follow-up had been completed. All of these patients had attained undetectable hCG values before the conception and in all cases had confirmation of a normal intrauterine pregnancy by ultrasound.
The following factors were evaluated as possible predictors of failure to complete hCG follow-up: Patients' age, gravidity, parity, marital status, insurance type, and distance from residence to trophoblast center. Because only 13 patients were lost to follow-up without attaining even one undetectable hCG level, this small group of patients did not lend itself to statistical analysis. We therefore included 42 patients who achieved weekly undetectable hCG values but did not begin monthly hCG follow-up and 67 patients who attained at least three undetectable weekly hCG values but did not complete monthly follow-up. The only factor associated with discontinuation of follow-up in these 122 patients was the distance from patient residence to trophoblast center. Women who lived farther than 20 miles from the New England Trophoblastic Disease Center versus those who lived closer were more likely to prematurely discontinue follow-up (P = .001). The percentages of patients who failed to complete follow-up as related to the distance from the trophoblastic center were as follows: distance from the trophoblastic center 0–20 miles, 26.1%; 21–100 miles, 56.5%; and farther than 100 miles, 50.0%. Patients' age, gravidity, parity, marital status, and insurance type were not significantly associated with failure to complete follow-up.
Importantly, among 320 patients who achieved at least one undetectable hCG level, no patient had any evidence of relapse of gestational trophoblastic tumor.
It is well recognized that complete and partial molar pregnancies have the potential of developing persistent gestational trophoblastic tumor. 1–5 Careful postevacuation hCG monitoring is therefore recommended for patients with molar pregnancy to assure the attainment of remission and to prevent the development of undetected persistent gestational trophoblastic tumor. Preevacuation hCG levels were significantly higher in complete versus partial molar pregnancies (P = .02), and the number of weeks until hCG became undetectable was significantly higher in complete moles (P = .001). These data were consistent with previously reported results. 16,17 Unfortunately, despite strong efforts of patient education and support, many patients with molar pregnancy do not complete hCG follow-up. Among the 333 patients who did not develop persistent gestational trophoblastic tumor, only 211 patients (63%) completed the recommended gonadotropin monitoring. Thirteen patients (4%) were lost to follow-up without achieving even one undetectable hCG value, and 109 patients (33%) attained undetectable hCG values but did not complete their entire follow-up. We decided to randomly select 400 patients with molar pregnancy for this study because it was felt that this number of patients would provide useful information concerning predictors of failure to complete hCG follow-up and the potential consequences of failing to complete follow-up. Our experience with postmolar hCG monitoring is similar to the published data from the University of Southern California, Korea, and Chicago. 13–15
We evaluated several potential factors that may be associated with failure to complete gonadotropin follow-up, including Patients' age, gravidity, parity, marital status, insurance type, and distance from residence to trophoblast center. Only the distance from the patient residence to the trophoblast center was significantly associated with the likelihood of not completing follow-up. Patients who lived farther than 20 miles from our center were more likely not to complete their hCG monitoring. Special efforts should therefore be made to encourage the completion of hCG monitoring in patients with molar pregnancy who live at a greater distance from the follow-up center.
Importantly, among the 320 patients who achieved at least one undetectable hCG level, none developed any evidence of relapse of gestational trophoblastic tumor. Prolonged hCG monitoring can be an economic, social, and emotional burden for patients with molar pregnancy. It would be helpful if other centers with considerable experience in observing patients with molar pregnancy reported their data concerning the frequency of relapse of gestational trophoblastic tumor in patients who achieve undetectable hCG levels. It appears that once a patient with molar pregnancy achieves undetectable hCG values, the risk of gestational trophoblastic tumor relapse is extraordinary low.
Patients with molar pregnancy are strongly encouraged to use reliable contraception and to complete recommended hCG follow-up to assure the attainment of remission and to minimize the risk of undetected gestational trophoblastic tumor relapse. However, it is possible that the interval of hCG monitoring of patients with molar pregnancy may be shortened without compromising patient health and safety. As more data are published concerning the risk of gestational trophoblastic tumor relapse in patients with molar pregnancy who achieve undetectable hCG levels, it may be appropriate to reevaluate current follow-up recommendations.
1. Kohorn EI. Hydatidiform mole and gestational trophoblastic disease in Southern Connecticut. Obstet Gynecol 1982;59:78–84.
2. Lurain JR, Brewer JI, Torok EE, Halpern B. Natural history of hydatidiform mole after primary evacuation. Am J Obstet Gynecol 1983;145:591–5.
3. Soper JT, Clarke-Pearson D, Hammond CB. Metastatic gestational trophoblastic disease: Prognostic factors in previously untreated patients. Obstet Gynecol 1988;71:338–43.
4. Bolis G, Belloni C, Bonazzi C, Mangili G, Presti M, Zanaboni F, et al. Analysis of 309 cases after hydatidiform mole: Different follow-up program according to biologic behavior. Tumori 1988;74:93–6.
5. Berkowitz RS, Goldstein DP. Chorionic tumors. N Engl J Med 1996;335:1740–8.
6. Seckl MJ, Fisher RA, Salerno G, Rees H, Paradinas FJ, Foskett M, et al. Choriocarcinoma and partial hydatidiform moles. Lancet 2000;356:36–9.
7. Bagshawe KD, Lawler SD, Paradinas FJ, Dent J, Brown P, Boxer GM. Gestational trophoblastic tumors following initial diagnosis of partial hydatidiform mole. Lancet 1990; 335:1074–6.
8. Garner EIO, Lipson E, Bernstein MR, Goldstein DP, Berkowitz RS. Subsequent pregnancy experience in patients with molar pregnancy and gestational trophoblastic tumor. J Reprod Med 2002;47:380–6.
9. Kim JH, Park DC, Bae SN, Namkoong SE, Kim SJ. Subsequent reproductive experience after treatment for gestational trophoblastic disease. Gynecol Oncol 1998;71:108–12.
10. Woolas RP, Bower M, Newlands ES, Seckl M, Short D, Holden L. Influence of chemotherapy for gestational trophoblastic disease on subsequent pregnancy outcome. Br J Obstet Gynaecol 1998;105:1032–5.
11. Ngan HY, Wong LC, Ma HK. Reproductive performance of patients with gestational trophoblastic disease in Hong Kong. Acta Obstet Gynecol Scand 1988;67:11–4.
12. American College of Obstetricians and Gynecologists. Management of gestational trophoblastic disease. ACOG technical bulletin no. 178. Washington: American College of Obstetricians and Gynecologists, 1993.
13. Schlaerth JB, Morrow CP, Kletzky OA, Nalick RH, D'Ablaing GA. Prognostic characteristics of serum human chorionic gonadotropin titer regression following molar pregnancy. Obstet Gynecol 1981;58:478–82.
14. Kim DS, Moon H, Kim KT, Moon YJ, Hwang YY. Effects of prophylactic chemotherapy for persistent trophoblastic disease in patients with complete hydatidiform mole. Obstet Gynecol 1986;67:690–4.
15. Massad LS, Abu-Rustum NR, Lee SS, Renta V. Poor compliance with postmolar surveillance and treatment protocols by indigent women. Obstet Gynecol 2000;96:940–4.
16. Paradinas FJ, Browne P, Fisher RA, Foskett M, Bagshawe KD, Newlands E. A clinical, histopathological and flow cytometric study of 149 complete moles, 146 partial moles and 107 non-molar hydropic abortions. Histopathology 1996;28:101–10.
17. Rob L, Rovova H, Pluta M, Kulovany E, Hrehorcak M, Chmel R, et al. Regression of hCG in various types of molar pregnancies—clinical course and prognosis. Ceska Gynecol 2001;66:230–5.